https://github.com/kali414/salesduo_assignemnt
AI-powered Meeting Summarizer API built with Python (Flask) and Node.js (Express), using Gemini 1.5 Flash to generate summaries, decisions, and action items from meeting transcripts.
https://github.com/kali414/salesduo_assignemnt
express flask nodejs python
Last synced: 3 months ago
JSON representation
AI-powered Meeting Summarizer API built with Python (Flask) and Node.js (Express), using Gemini 1.5 Flash to generate summaries, decisions, and action items from meeting transcripts.
- Host: GitHub
- URL: https://github.com/kali414/salesduo_assignemnt
- Owner: Kali414
- License: mit
- Created: 2025-05-31T11:40:35.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-06-01T02:26:21.000Z (about 1 year ago)
- Last Synced: 2025-08-22T08:23:13.733Z (11 months ago)
- Topics: express, flask, nodejs, python
- Language: JavaScript
- Homepage:
- Size: 41 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 📝 Meeting Summarizer API (Python + Node.js)
This repository contains two separate implementations of a **Meeting Summarizer API** that extracts structured insights from meeting transcripts using **Google Gemini API**.
- 🐍 Python (Flask)
- 🟢 Node.js (Express)
Both versions leverage **Google Gemini 1.5 Flash** to extract structured insights from meeting transcripts:
- ✅ A concise summary
- ✅ Key decisions made
- ✅ Action items (task, owner, deadline)
---
## 📂 Project Structure
- /python → Python Flask Langchain implementation
- /node → Node.js Express implementation
- /data → Sample input transcript files
```yaml
Each implementation includes its own:
- API setup
- Route to accept meeting transcript
- Gemini API integration
- Sample output
---
## ✨ Features
- Accepts transcript input in:
- 🧾 Raw text
- 🧠 JSON body
- 📁 Uploaded `.txt` file
- Uses **Gemini 1.5 Flash** model for summarization
- Returns clean and structured **JSON output**
- API documentation & Postman-ready examples
---
## 📊 Sample Output Format
```json
{
"summary": "The team discussed project deadlines and assigned tasks.",
"decisions": [
"Final review scheduled for Sept 15"
],
"actionItems": [
{
"task": "Create slide deck",
"owner": "Alex",
"due": "Sept 10"
}
]
}
```
## 📦 Install & Run
Each implementation has its **own README** with complete installation and usage instructions.
---
## 📦 Setup Instructions
> 🛠️ Follow the setup and run instructions in the respective folder’s README:
- [Python Flask Setup Guide →](./python/README.md)
- [Node.js Express Setup Guide →](./node/README.md)
Each README includes:
- Installation steps
- Running the local development server
- Testing the `/process_meeting` API route using Postman or curl
---
## 🧪 Example Datasets
Two sample transcript files are included under the /data directory:
example1.txt — Sample meeting transcript with clear responsibilities
example2.txt — A longer, real-world style meeting log
data_json.txt - Use the content of the file , if the transcript data is already structured in JSON format.
You can use these files to test the API by uploading via Postman or passing the text as input.